Generating Creative Chess Puzzles

2025-10-27

AI Summary

Generative AI struggles to produce highly creative and strongly counterintuitive chess puzzles. Method: This paper proposes a novel generative framework integrating reinforcement learning with statistical analysis from chess engines. A policy network is initialized via supervised learning, and a multi-objective reward function—explicitly modeling “counterintuitiveness,” “aesthetic structure,” and “solution uniqueness”—is designed using statistical features derived from deep engine search (e.g., Stockfish). Contribution/Results: The reward mechanism improves counterintuitive puzzle generation efficiency by 10×; the proportion of counterintuitive puzzles rises significantly from 0.22% to 2.5%, surpassing human expert–curated puzzle datasets. Independently evaluated by three world champion–level players, the generated puzzles were selected for an anthology of high-creativity AI puzzles, demonstrating dual breakthroughs in authenticity and artistic quality.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Search and Optimization: Adversarial Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

While Generative AI rapidly advances in various domains, generating truly creative, aesthetic, and counter-intuitive outputs remains a challenge. This paper presents an approach to tackle these difficulties in the domain of chess puzzles. We start by benchmarking Generative AI architectures, and then introduce an RL framework with novel rewards based on chess engine search statistics to overcome some of those shortcomings. The rewards are designed to enhance a puzzle's uniqueness, counter-intuitiveness, diversity, and realism. Our RL approach dramatically increases counter-intuitive puzzle generation by 10x, from 0.22% (supervised) to 2.5%, surpassing existing dataset rates (2.1%) and the best Lichess-trained model (0.4%). Our puzzles meet novelty and diversity benchmarks, retain aesthetic themes, and are rated by human experts as more creative, enjoyable, and counter-intuitive than composed book puzzles, even approaching classic compositions. Our final outcome is a curated booklet of these AI-generated puzzles, which is acknowledged for creativity by three world-renowned experts.

Problem

• Generating creative and counter-intuitive chess puzzles
• Overcoming limitations in AI puzzle uniqueness and diversity
• Enhancing aesthetic quality and realism in puzzle generation

Innovation

• Reinforcement learning framework with novel reward design
• Rewards based on chess engine search statistics
• Enhancing puzzle uniqueness and counter-intuitiveness